aqibsaeed / research-paper-categorization

Research paper classification using machine learning and NLP

Research paper categorization

• Python notebook and dataset for the blog post.
• Make sure you have Python 2.7 and following libraries installed: scikit-learn (0.17), pandas, numpy, nltk and gensim. Anaconda distribution is recommended.

research-paper-categorization's Issues

Python2 or Python 3...xrange or range

Hello,

When I run the code, the first error I get is that range i snot found as I am running the code in python3. Should this be run in python2? I tried it sklearn was not found.

def preProcessing(features):
num_titles = features.size
clean_titles = []
stops = set(stopwords.words("english"))
for i in range( 0, num_titles): < problem is here
#letters_only = re.sub("[^a-zA-Z]", " ", features[i])
words = features[i].lower().split()
words = [w.lower() for w in words if not w in stops]
clean_titles.append(" ".join(words))
return clean_titles

TypeError: only integer arrays with one element can be converted to an index

Hello,

Thanks for sharing this project. I am trying to run it and I have encountered a couple of problems. I will submit another pug report in a moment.

Here I am getting an error that I tried to fix by setting the inputs to document_term_matrix as numpy arrays but that still do not help. Do you have any ideas on what the problem is?

I am using Python 3.

Regards

Burke

TypeError Traceback (most recent call last)
in ()
----> 1 precision, recall, fscore = crossValidate(chisqDtm, labels, "SVM", 10)

in crossValidate(document_term_matrix, labels, classifier, nfold)
16 for train_index, test_index in skf:
17 print(train_index, test_index)
---> 18 X_train, X_test = document_term_matrix[train_index], document_term_matrix[test_index]
19 y_train, y_test = labels[train_index], labels[test_index]
20 model = clf.fit(X_train, y_train)

TypeError: only integer arrays with one element can be converted to an index